A Comparison Across Decay Classes and Tree Species
Total Page：16
File Type：pdf, Size：1020Kb
United States Department of DDifferencesifferences BBetweenetween Agriculture Forest Service SStandingtanding aandnd DDownedowned Northern Research Station DDeadead TTreeree WWoodood DDensityensity Research Paper NRS-15 RReductioneduction FFactors:actors: A CComparisonomparison AAcrosscross DDecayecay CClasseslasses aandnd TTreeree SSpeciespecies MMarkark EE.. HHarmonarmon CChristopherhristopher WW.. WWoodalloodall BBeckyecky FFasthasth JJayay SSextonexton MMishaisha YYatkovatkov Abstract Woody detritus or dead wood is an important part of forest ecosystems and has now become a routine facet of forest monitoring and inventory. Biomass and carbon estimates of dead wood depend on knowledge of species- and decay class- specifi c density or density reduction factors. While some progress has been made in determining these parameters for dead and downed trees (DD), there are very few estimates of these key parameters for standing dead trees (SD). In this study we evaluated indicators of decay to relate subjective SD and DD decay classifi cations then compared SD and DD density and density reduction factors by decay class for a total of 19 tree species at nine sites in the United States and Russia. A set of preliminary decay reduction factors for SD trees by species and decay class was developed for tree species inventoried by the U.S. Department of Agriculture’s Forest Inventory and Analysis (FIA) program. Results indicate that SD density declined with decay class for all examined species. For six of the examined species, SD tree density could be assumed equal to that of DD tree density. The most common situation (13 species) was for SD density to be higher than for DD density. The likely cause of these differences was the drier microenvironment of SD which slows decomposition relative to that of DD. By applying these results, a new set of SD density reduction factors was developed for 260 species inventoried by FIA in forests of the United States. Comparison of biomass estimates using this study’s proposed SD density reduction factors with existing SD volume estimates based solely on undecayed wood density indicated a possible biomass overestimate of 16.5 percent for Minnesota. Given the size of the potential biases involved and the uncertainty associated with decay reduction factors developed in this study, additional work to develop a multispecies SD decay class system and further empirical sampling of SD tree density is recommended. Cover Photo Standing dead tree on the shores of Octopus Lake, Boundary Waters Canoe Area Wilderness, Superior National Forest, Minnesota, USA. Christopher W. Woodall, U.S. Forest Service. Manuscript received for publication 19 April 2011 Published by: For additional copies: U.S. FOREST SERVICE U.S. Forest Service 11 CAMPUS BLVD SUITE 200 Publications Distribution NEWTOWN SQUARE PA 19073 359 Main Road Delaware, OH 43015-8640 Fax: (740)368-0152 August 2011 Email: [email protected] Visit our homepage at: http://www.nrs.fs.fed.us/ INTRODUCTION Woody detritus or dead wood is an important part of forest ecosystems, providing habitats and food sources, a store of carbon, nutrients, and water, as well as infl uencing geomorphic processes (Franklin et al. 1987, Harmon et al. 1986, Triska and Cromack 1980). In particular, standing dead trees provide structural complexity and enhance biodiversity in many forest ecosystems (Jonsell et al. 1998, Kruys et al. 1999, Nilsson et al. 2001). Th erefore, dead wood is increasingly being included in ecological studies and environmental assessments (Woodall et al. 2009), with the number of publications on this subject rapidly growing. Both downed (DD) and standing dead (SD) wood store carbon and are important components of forest carbon dynamics, Th e Authors but also release carbon via decomposition and combustion in fi res. MARK E. HARMON is a Richardson However, given the contrasting microenvironments that SD and DD Chair and professor in forest science with the trees inhabit, their decay dynamics and attributes may vary (Aakala Department of Forest Science, Oregon State et al. 2008, Aakala 2010, Lee 1998, Vanderwel et al. 2006). Accurate University, Corvallis, OR. estimates of standing dead tree carbon stocks are important for monitoring carbon fl ux, for characterizing processes such as respiration CHRISTOPHER W. WOODALL is research and combustion losses, and for validating simulation models that are forester with the U.S. Forest Service, Northern used to project how carbon dynamics will change in the future. Research Station’s Forest Inventory and Analysis unit, St, Paul MN. It is diffi cult to develop accurate estimates of dead wood biomass for several reasons. First, until recently there have been few systematic, BECKY FASTH is a senior research assistant regional-scale dead wood inventories, resulting in highly uncertain with the Department of Forest Science, Oregon dead wood population estimates (Harmon et al. 2001). However, State University, Corvallis, OR. regional-scale inventories of dead wood are becoming more common (Woodall et al. 2009). For example, the Forest Inventory and Analysis JAY SEXTON is a senior research assistant (FIA) program of the U.S. Forest Service now conducts a national with the Department of Forest Science, Oregon inventory of woody material including fi ne woody detritus (FWD) State University, Corvallis, OR. and DD (for more details see Woodall and Monleon 2008). Second, given the size and actively decomposing nature of dead trees, direct MISHA YATKOV is a Ph.D. candidate with measurements of biomass during forest inventories are not possible. the Department of Forest Science, Oregon Instead, the volume of individual DD pieces is estimated from fi xed- State University, Corvallis, OR. area plot or line-intersect transect estimators (Woodall and Monleon 2008). Biomass of DD is then often estimated by multiplying volume by an estimate of density specifi c to a dead wood piece’s individual attributes (e.g., species and stage of decay). Density reduction factors may be defi ned as the ratio of the decayed density (current mass/volume) of a piece of dead wood compared to its undecayed density (initial live tree mass/volume; Miles and Smith 2009). Th e development of DD species- and decay class-specifi c density and decay reduction values for the FIA program has off ered the promise of substantially reducing the uncertainty associated with DD mass/carbon population estimates resulting from large-scale inventories (Harmon et al. 2008). Unfortunately, there are very few decay class- or species- 1 specifi c estimates of density or decay reduction factors for temperate forests of the northern hemisphere. As the SD trees. Th e lack of these estimates is likely due to the decay classifi cation system for both SD and DD trees are danger in felling snags during sampling and the historical qualitative (i.e., visual assessments of wood condition lack of emphasis on SD biomass/carbon estimates during and structure) and separate, the two classifi cation systems traditional forest inventories focused on estimating live need to be linked in order to propose SD to DD wood tree growing stock volume on timberlands. density relationships by decay class. First, indicators of decay were evaluated for SD and DD decay classes Several assumptions regarding wood density infl uence to ensure decay classifi cation system compatibility. biomass estimates of SD trees. In the case of the FIA Second, the density of SD and DD was compared for program, it is currently assumed that the density of the same decay classes and species. Th ese ratios were SD trees is the same as that of live trees. While this then used to estimate SD density and density reduction seems unlikely, in some situations (e.g., extremely slow factors for species that have not been sampled across the decomposition in arid environments) this assumption may United States. Finally, to initially assess the impact of be accurate. Alternately, (e.g., Smithwick et al. 2002) the wood density decay reduction factors on SD population density of SD trees for a given decay class may be assumed estimates, a SD tree inventory in Minnesota was used to be similar to that of DD wood. While this would seem to calculate state SD biomass using no decay reduction to be a more realistic assumption, it has yet to be tested factor and using factors developed in previous objectives. across a wide range of tree species and environments. Th e uncertainty associated with SD tree density is likely MATERIALS AND METHODS to have major impacts on biomass, carbon, and fuel Study Areas estimates, but without knowledge of how SD tree wood density changes through its decay process, the eff ect of To ensure our study acquired a robust estimate of SD this uncertainty cannot be quantifi ed. to DD, we sampled fi ve study sites in the United States and four in Russia representing a range of species and OBJECTIVES climates in North America and Russia (Table 1). Given the similarity between the temperate and boreal forests of Given that deadwood inventories are rapidly becoming North America and Russia, it was felt that the additional a common component of forest resource inventories statistical rigor resulting from the combination of data around the world, there is a need to refi ne the constants outweighed the loss of some regional specifi city. As study used in biomass estimation procedures for both SD and objectives focus on tree species in the United States along DD. Our objective was to determine SD species- and with greater diversity among the study sites (e.g., species decay-class specifi c density and density reduction factors composition), study site locations in the United States for forest tree species in the United States by linking were kept separate and identifi ed by unique abbreviations empirical SD data from a subset of study species with in subsequent tables, while Russian study sites were that of DD of those same species to develop constants pooled with the abbreviation “RUS.” Site descriptions for that can then be applied to other SD species.